Summary form only given. A neural network scheme is proposed to solve the inverse kinematic problem for redundant robots in an environment with or without obstacles. The inverse kinematic solution of a four link planar robot is simulated using a multilayer feedforward network with hidden units having sigmoidal functions and output units having linear functions. The results show that the proposed scheme provides very satisfactory solutions
Published in:
Robotics and Automation, 1993. Proceedings., 1993 IEEE International Conference on
Date of Conference: 2-6 May 1993